staging: greybus: light: Release memory obtained by kasprintf

Free memory region, if gb_lights_channel_config is not successful.

Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Reviewed-by: Rui Miguel Silva <rmfrfs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: media: atomisp: Use kmalloc_array instead of kmalloc

Prefer kmalloc_array over kmalloc with multiply
because kmalloc_array performs additional checks
before memory allocation.

Fix the following issue reported by checkpatch.pl:
* WARNING: Prefer kmalloc_array over kmalloc with multiply

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

atomisp2: remove cast from memory allocation

Patch removes the following warning issued was coccicheck:
WARNING: casting value returned by memory allocation function to (char *) is
useless.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: media: atomisp: pci: Place constant on the right side in comparissons

Fix issue found by checkpatch.pl script.
WARNING: Comparisons should place the constant on the right side of the test

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: media: atomisp: pci: Move open brace '{' on the next line

Fix the following issue found by checkpatch.pl:
ERROR: open brace '{' following function definitions go on the next line

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: media: atomisp: Add blank line after declarations

This patch adds a blank line after declarations to
improve code readability.

Issue find by checkpatch.pl script.
WARNING: Missing a blank line after declarations

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: media: atomisp: Use unsigned int instead of unsigned

Fix the checkpatch.pl issue:
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8188eu: core: Fix line over 80 characters

Fix warning reported by checkpatch.pl script:
WARNING: line over 80 characters

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8188eu: core: Add spaces around '+'

Improve the coding style by adding spaces around
arithmetic operation.

Issue reported by checkpatch.pl script.

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8188eu: core: Use __func__ instead of function name

Replace the function name from format string with the
constant __func__ to avoid multiple changes in case
the name of the function will be modified.

Issue reported by checkpatch.pl script.

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: wrap lines in 80 characters

Perform cleanup for all function declarations in core/rtw_mlme_ext wherever
checkpatch complains about lines being over 80 characters long.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: remove implicit bool->int conversions

Implicit type conversions are bad; they hinder readability of code and have
potential to cause bugs. Here the variable wait_ack is always supplied a bool
value while in function declarations it is defined as an int type. Fix it by
defining wait_ack a bool type in all usages.

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: remove unneeded conversions to bool

Patch suppresses the following warning issued by coccicheck:
WARNING: conversion to bool not needed here

Signed-off-by: Aishwarya Pant <aishpant@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Remove typedef struct vchiq_2835_state_struct

This patch removes typedef from struct and renames it from "typedef
struct vchiq_2835_state_struct" to "struct vchiq_2835_state" as per
kernel coding standards.

Signed-off-by: Mihaela Muraru <mihaela.muraru21@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: Remove extern variable

This patch removes extern variable vchiq_arm_log_level, because it is
already declared in vchiq_arm.h

Issue found by checkpatch.pl

Signed-off-by: Mihaela Muraru <mihaela.muraru21@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: vc04_services: remove unused variables

the volatile fields of bcm2835_alsa_stream -
control and status are not used.

$ grep bcm2835_alsa_stream
>From the above command all instances we see that
all variables of 'bcm2835_alsa_stream' are declared as
'alsa_stream'

So search for 'control' wherever we have 'alsa_stream'

$ grep -l 'alsa_stream' | xargs grep "control"
The above command returns where we don't any usage of
'control' field. which means that there is no usage of
these fields.

similarly for 'status' we see no usages.

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: vc04_services: bcm2835-camera: use BIT macro

Use BIT macro instead of left shift

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Acked-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: fix typos

Some words are misspelled. Fix typos.

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: please do not use multiple blank lines

Blank lines use up extra space in file and makes the file
larger. So do not use multiple blanklines

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: fsl-mc: remove unnecessary parenthesis

Remove unnecessary parenthesis

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: pi433: fix grammar issues

- This commit fixes spelling issues in documentation files.

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: irda: drivers: Replace seq_printf with seq_puts or seq_putc

Replace seq_printf with seq_puts or seq_putc when
there is no argument list.

Fix the checkpatch warning:
WARNING: Prefer seq_puts to seq_printf

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: lustre: lnet: replace list_for_each with list_for_each_entry

Replace use of the combination of list_for_each() and list_entry() with
list_for_each_entry() to simplify the code and remove variables that are
used only in list_for_each().
Issue found and corrected using Coccinelle script:

@r@
expression head, member, e;
type T1, T2, T3;
iterator name list_for_each, list_for_each_entry;
identifier pos, var;
@@

-T1 *pos;
...when!=pos

-list_for_each(pos, head)
+list_for_each_entry(var, head, member)
{
...when!=pos
   when!=T3 *var;
-var = list_entry(pos, T2, member);
...when!=pos
}
...when!=pos

Signed-off-by: Haneen Mohammed <hamohammed.sa@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: Blank lines aren't necessary before a close brace '}'

Before closing the brace, the blank line is not needed. That blank
line has not meaning here. so remove it.

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: wilc1000: No space is necessary after a cast

A space after type casting is not needed. All the other typecasts
in this file don't put space after typecast. so remove space
after typecast at this line.

checked with checkpatch.pl script

Signed-off-by: Keerthi Reddy <keerthigd4990@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Fix comment block coding style

This is a patch to the spk_ttyio.c file that fix up a comment block
warninig, found by checkpatch.pl tool, by adding */ on a separte line.

WARNING: Block comments use a trailing */ on a separate line

Signed-off-by: Mihaela Muraru <mihaela.muraru21@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: speakup: Use octal permissions '0444'

Fixed the following checkpatch warning:

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal
permissions '0444'.

Signed-off-by: Mihaela Muraru <mihaela.muraru21@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visornic: clean up parenthesis

Clean up unneeded parenthesis reported by checkpatch.pl.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorhba: clean up parenthesis

Clean up unneeded parenthesis reported by checkpatch.pl.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: Fix extraneous parenthesis

Clean up unneeded () that are being reported by checkpatch.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: fix comments in vbuschannel.h

This patch fixes some comments in the file vbuschannel.h to make
it more uniform.

Signed-off-by: Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: fix comments and spacing issues in visorbus.h

This patch fixes some comments and spacings in the file visorbus.h
to make it more uniform.

Signed-off-by: Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: fix line over 80 characters checkpatch warning

This patch fixes two checkpatch issues of line over 80 characters
in visorchipset.c.

Signed-off-by: Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: simplify visorchannel_create_guts

Removing the two wrapper functions dealing with visorchannel_create() and
instead just always use a new version of visorchannel_create() with an
additional parameter.

Signed-off-by: Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: check the whole channel instead of just guid for match

Validate that the channel contents match the channel type that we are
matching.

Signed-off-by: Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: Remove unused enum

Removes unused enum that pertains to s-Par diagnostics capabilities.

Signed-off-by: David Binder <david.binder@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: Remove unnecessary macro to generate signature

Removes unnecessary SIGNATURE_16 macro used to generate a simple 32-bit
signature.

Signed-off-by: David Binder <david.binder@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: Rename channel.h to visorchannel.h

In order to avoid a potential conflict with other drivers that define a
channel.h file, the s-Par channel.h file is renamed in accordance with the
surrounding driver-set namespace.

Signed-off-by: David Binder <david.binder@unisys.com>
Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: clean up parameter formatting

Cleans up parameter formatting in visorbus.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: shrink name of POLLJIFFIES_CONTROLVMCHANNEL_XXX

Shrinks the name of POLLJIFFIES_CONTORLVMCHANNEL_FAST and
POLLJIFFIES_CONTROLVMCHANNEL_SLOW to just POLLJIFFIES_CONTROLVM_XXXX.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorchipset.c: Remove useless initializations

Found several more useless initializations in visorchipset.c get rid of
them.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: remove visorbus_type extern

The variable visorbus_type could be contained to the visorbus_main.c
file by moving the two functions that referenced it visorchipset.c. This
allowed us to remove the incorrect extern from the include file visorbus.h.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: keep the success path on the left side

If you can't find the dev, it is an error. Indent for the error, instead
of the success.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: consolidate a comment

Use all 80 characters of the line to consolidate a comment.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: remove signature_32

Remove signature_32 since it was only being referenced in one location.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: reorder to avoid forward declaration

The include file visorbus.h has a forward declaration of visor_device.
A simple reorder of the file removed the need for the forward
declaration.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: vbuschannel doesn't use ctypes

Don't include ctypes in vbuschannel.h, it isn't used by ctypes, but it is
used by visorbus_main.c so include it there.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: include: remove io.h from channel.h

Channel.h was including io.h and not using it. Remove the #include.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: only include a file where it is used.

The header file visorbus.h included several linux headers that were
used by the source files that include it. Move the includes to the
files that actually use them.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: to_visor_device/to_visor_driver move up

The macros to_visor_device and to_visor_driver are fairly small; move them
up to the declaration block to save some space.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorbus_main.c: remove blank lines

Removes several extraneous blank lines in visorbus_main.c.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorchipset: clean up blank lines in visorchipset

Remove extraneous blank lines and get consistency of blank lines.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: fix newlines for visorchannel

Remove extraneous blank lines and add some lines that are needed. General
philosophy -- declaration, space, pre-conditions (if needed), space, then
rest of function.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: Remove unneeded initialization

No need to set drv to NULL.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: simplify debugfs print statment

Simplify debugfs statement so it is clearer and more compact.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: clean up formatting of function definitions

Consolidate function parameters onto the same line if they can fit.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: shrink the name of a variable

The field debugfs_client_bus_info can be reduced to debugfs_bus_info.
The word client is extraneous in this context.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: keep the success path on the left

The code was indenting for the successful path and then combining the
error and success path for the rest of the function. Correct it so the
success path is not indented.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: clean up initializations

Don't compute value of offset during initialization. Assigning a value to
offset should happen after we have verified all of its components.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys; visorbus: visorbus_main.c: remove extraneous newline

Use all 80 characters when calling functions.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: fix up device_destroy

Visorbus_remove_instance calls device_unregister which in turn will call
visorbus_release_busdevice. The function visorbus_release_busdevice was
freeing the visor_device. The code in visorbus_remove_instance was then
trying to reference the visor_device. This patch cleans up the code so
the visor_device is not referenced after the unregister.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: NULL pending_msg_hdr not an error

A NULL pending_msg_hdr is not an error, it just means that the firmware
does not want an error response for that message.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: don't put code in declaration.

Simplify the code by not putting the assignment in the declaration. Define
it and then assign it, so the function doesn't get bunched up on the
right.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: use guid_is_equal instead of memcmp

The function publish_vbus_dev_info was doing a memcmp of guids. It should
be using the kernel provided guid_equal function instead.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorbus_main.c: cleanup and consolidate comments

Use all 80 characters of the line for comments to help reduce the number
of extra lines in the code.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorbus_private.h remove extra blank lines

Remvoe several unneeded blank lines.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: fix alignment of paranethesis

Correct indenting of parameters when calling the functions in the file
visorchipset.c.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys; visorbus: visorchipset.c: remove extraneous newline

Use all 80 characters when calling functions.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: visorchannel: use 80 characters for comments

In the file visorhcannel.c use all 80 characters for comments instead of
shortening them.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: clean up parser_string_get function

The function parser_string_get was writing its own strnlen function, use
the kernel version instead.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: don't need local variable cmd

The local variable cmd was just being used to dereference a field inside
of it. Get rid of the variable and just derefernce the parameter being
passed in.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: unisys: visorbus: simplify commenting in controlvm_init

During the process of udpating the features bits, put a block comment for
all updates instead of a separate comment for each one.

Signed-off-by: David Kershner <david.kershner@unisys.com>
Reviewed-by: Tim Sell <timothy.sell@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: Fix indent coding style issue

This patch fixes an indentation coding style issue found by
checkpatch.pl:

WARNING: suspect code indent for conditional statements (16, 32)

Signed-off-by: Mihaela Muraru <mihaela.muraru21@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8723bs: core: Remove boolean comparison

Remove comparison to bool in order to improve
the clearness of the code.

Issue found using coccinelle script.

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

iio:stm32-lp-timer and ep93xx: drop assign iio_info.driver_module and iio_trigger_ops.owner

The equivalent of both of these are now done via macro magic when
the relevant register calls are made.  The actual structure
elements have gone away.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'iio-for-4.15a' of git://git./linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

Round one of new device support, features and cleanup for IIO in the 4.15 cycle.

Note there is a misc driver drop in here given we have support
in IIO and the feeling is no one will care.

A large part of this series is a boiler plate removal series avoiding
the need to explicitly provide THIS_MODULE in various locations.
It's very dull but touches all drivers.

New device support
* ad5446
  - add ids to support compatible parts DAC081S101, DAC101S101,
    DAC121S101.
  - add the dac7512 id and drop the misc driver as feeling is no
    one is using it (was introduced for a board that is long obsolete)
* mt6577
  - add bindings for mt2712 which is fully compatible with other
    supported parts.
* st_pressure
  - add support for LPS33HW and LPS35HW with bindings (ids mostly).

New features
* ccs811
  - Add support for the data ready trigger.
* mma8452
  - remove artifical restriction on supporting multiple event types
    at the same time.
* tcs3472
  - support out of threshold events

Core and tree wide cleanup
* Use macro magic to remove the need to provide THIS_MODULE as part of
  struct iio_info or struct iio_trigger_ops.  This is similar to
  work done in a number of other subsystems (e.g. i2c, spi).

  All drivers are fixed and then the fields in these structures are
  removed.

  This will cause build failures for out of tree drivers and any
  new drivers that cross with this work going into the kernel.

  Note mostly done with a coccinelle patch, included in the series
  on the mailing list but not merged as the fields no longer exist
  in the structures so the any hold outs will cause a build failure.

Cleanups
* ads1015
  - avoid writing config register when it doesn't change.
  - add 10% to conversion wait time as it seems it is sometimes
    a little small.
* ade7753
  - replace use of core mlock with a local lock.  This is part of a
    long term effort to make the use of mlock opaque and single
    purpose.
* ade7759
  - expand the use of buf_lock to cover previous mlock cases.  This
    is a slightly nicer solution to the same issue as in ade7753.
* cros_ec
  - drop an unused variable
* inv_mpu6050
  - add a missing break in a switch for consistency - not actual
    bug,
  - make some local arrays static to save on object code size.
* max5481
  - drop manual setting of the spi module owner as handled by the
    spi core.
* max5487
  - drop manual setting of the spi module owner as handled by the
    spi core.
* max9611
  - drop explicit setting of the i2c module owner as handled by
    the i2c core.
* mcp320x
  - speed up reads on single channel devices,
  - drop unused of_device_id data elements,
  - document the struct mcp320x,
  - improve binding docs to reflect restrictions on spi setup and
    to make it explicit that the reference regulator is needed.
* mma8452
  - symbolic to octal permissions,
  - unsigned to unsigned int.
* st_lsm6dsx
  - avoid setting odr values multiple times,
  - drop config of LIR as it is only ever set to the existing
    defaults,
  - drop rounding configuration as it only ever matches the defaults.
* ti-ads8688
  - drop manual setting of the spi module owner as handled by the
    spi core.
* tsl2x7x
  - constify the i2c_device_id,
  - cleanup limit checks to avoid static checker warnings (and generally
    have nicer code).

iio: imu: st_lsm6dsx: remove rounding configuration

Remove rounding configuration since it is enabled by default for the
FIFO output registers

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: imu: st_lsm6dsx: remove LIR configuration

Remove Latched Interrupt configuration since it is enabled by default
for FIFO watermark interrupt

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: mt2712: Add compatible node for mt2712.

This commit adds mt2712 compatible node.

Signed-off-by: Zhiyong Tao <zhiyong.tao@mediatek.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

dt-bindings: adc: mt2712: add binding documention

The commit adds mt2712 compatible node in binding document.

Signed-off-by: Zhiyong Tao <zhiyong.tao@mediatek.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

dt-bindings: iio: adc: mcp320x: Update for mcp3550/1/3

All chips supported by this driver clock data out on the falling edge
and latch data in on the rising edge, hence SPI mode (0,0) or (1,1)
must be used.

Furthermore, none of the chips has an internal reference voltage
regulator, so an external supply is always required and needs to be
specified in the device tree lest the IIO "scale" in sysfs cannot be
calculated.

Document these requirements in the device tree binding, add compatible
strings for the newly supported mcp3550/1/3 and explain that SPI mode
(0,0) should be preferred for these chips.

Cc: Mathias Duckeck <m.duckeck@kunbus.de>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: imu: st_lsm6dsx: set sensor->odr value just in st_lsm6dsx_write_raw()

Update odr value in st_lsm6dsx_sensor data structure just in
st_lsm6dsx_write_raw() in order to avoid to set the same value
multiple times

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

Staging: iio: meter: ade7759: update locking method for write frequency

iio_dev->mlock is to be used only by the IIO core for protecting
device mode changes between INDIO_DIRECT and INDIO_BUFFER.

This patch replaces the use of mlock with the already established
buf_lock mutex.

Introducing an 'unlocked' spi_write_reg_16 function to be used by
ade7759_write_frequency avoids nested locks and maintains atomicity
between bus and device frequency changes.

Based on the solution found in ade7754 patch here:
https://marc.info/?l=linux-iio&m=149086659008991&w=2

Signed-off-by: Katie Dunne <kdunne@mail.ccsf.edu>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: ade7753: replace mlock with driver private lock

Replace driver usage of mlock with driver private lock to meet the new
model where usage of iio_dev->mlock is being redefined as protecting
operating mode changes(changes between BUFFER* and DIRECT modes).

Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: accel: mma8452: Fix code style warning for unsigned int declarations

Replace 'unsigned' with 'unsigned int'
to improve code readability.

Issue found by checkpatch.

Signed-off-by: Harinath Nampally <harinath922@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: accel: mma8452: Fix code style warning

Replace symbolic permissions with octal permissions
to improve code readability.

Issue found by checkpatch.

Signed-off-by: Harinath Nampally <harinath922@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: cros_ec: Remove unused variable

Fix gcc warning:
cros_ec_baro.c:130:25: warning: variable ‘ec_device’ set but not used

Signed-off-by: Paolo Cretaro <paolocretaro@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: iio: tsl2x7x: clean up limit checks

The background of this code is that we can either use the default
tables or load our own table with sysfs.  The default tables are three
element arrays of struct tsl2x7x_lux.  If we load the table with sysfs
then we can have as many as nine elements.  Which ever way we do it, the
last element is always zeroed out.

The most interesting part of this patch is in the
in_illuminance0_lux_table_show() function.  We were using the wrong
limit, "TSL2X7X_MAX_LUX_TABLE_SIZE * 3", when it should have been just
"TSL2X7X_MAX_LUX_TABLE_SIZE".  This creates a static checker warning
that we are going of bounds.  However, since the last element is
always zeroed out, that means we hit the break statement and the code
works correctly despite the wrong limit check.

I made several related readability changes.  The most notable that I
changed the MAX_DEFAULT_TABLE_BYTES define which was:

I renamed the define to TSL2X7X_DEFAULT_TABLE_BYTES because it's not the
max size, it's the only size.  Also the size should really be expressed
as sizeof(struct tsl2x7x_lux) * 3.  In other words, 12 * 3 instead of
4 * 9.  It's 36 bytes either way, so this doesn't change the behavior.

Finally, I created the TSL2X7X_DEF_LUX_TABLE_SZ define instead of using
the magic number 3.  I declared the default tables using that define
to hopefully signal to future programmers that if they want to use a
different size they have to update all the related code.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: mcp320x: Document struct mcp320x

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: mcp320x: Drop unnecessary of_device_id attributes

The driver sets a .data pointer for each .compatible string but never
calls of_device_get_match_data().  Instead, ADC properties are looked up
with spi_get_device_id().  The .data pointer is therefore unnecessary,
so drop it.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: adc: mcp320x: Speed up readout of single-channel ADCs

Single-channel converters such as mcp3001, mcp3201, mcp3301 and the
upcoming mcp3550/1/3 lack a MOSI pin, so there's no need to call
mcp320x_channel_to_tx_data() for them.

Moreover, instead of calling spi_read() for these converters, which
generates an spi_message and spi_transfer on the stack on every readout,
it's more efficient to use the spi_message and spi_transfer[] included
in struct mcp320x (as we do for multi-channel ADCs), but initialize the
spi_message only with the receive transfer.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

iio: chemical: ccs811: Add support for data ready trigger

Add data ready trigger for hardware interrupts that signal
new, available measurement samples.

Cc: Daniel Baluta <daniel.baluta@gmail.com>
Cc: Alison Schofield <amsfield22@gmail.com>
Signed-off-by: Narcisa Ana Maria Vasile <narcisaanamaria12@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

staging: xgifb: make const array static to shink object code size

Don't populate const array LCDARefreshIndex on the stack, instead
make it static.  Makes the object code smaller by 340 bytes:

Before:
   text    data     bss     dec     hex filename
  84949   12336       0   97285   17c05 drivers/staging/xgifb/vb_setmode.o

After:
   text    data     bss     dec     hex filename
  84506   12432       0   96938   17aaa drivers/staging/xgifb/vb_setmode.o

(gcc version 7.2.0 x86_64)

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vc04_services: remove BCM2835_VCHIQ_SUPPORT_MEMDUMP

BCM2835_VCHIQ_SUPPORT_MEMDUMP lets you look through any user memory.
That's too big of an information leak from a security perspective.  The
debugging dumps need to be more specific to this driver.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:rtl8188eu:hal Fix wrong comparison to False

This patch solves the warning "Using comparison to false is error prone"

Signed-off-by: Janani Sankara Babu <jananis37@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging:rtl8188eu Fix remove semicolon in do {}while(0)

This patch removes the semicolon at the end of while statement in the
do while macro , inorder to avoid it behaving like compound statement.

Signed-off-by: Janani Sankara Babu <jananis37@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging/comedi/das16: Make timer initialization unconditional

With timer initialization made unconditional, there is no reason to
make del_timer_sync() calls conditionally, there by removing the test
of the .data field.

Cc: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: irda: net: Fix style issues

Fix minor coding style issues found by checkpatch.pl.

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: irda: net: Use NOT operator instead of comparison to NULL

Fix issues find by checkpatch.pl.
CHECK: Comparison to NULL could be written "!dongles"
CHECK: Comparison to NULL could be written "!tasks"

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: irda: net: Do not initialise statics to NULL

There is no need to initialize static variables to NULL
because they are stored in .bss segment. This segment
is initialized to 0 at the beginning of the code execution.

Issue found by checkpatch.pl.
ERROR: do not initialise statics to NULL

Signed-off-by: Georgiana Chelu <georgiana.chelu93@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>